The career goals of Dr. Shannon Carson include becoming a productive independent investigator who will conduct clinical research and outcomes studies involving critically ill patients. To gain the necessary background, he will participate in clinical research conferences and seminars at the UNC School of Medicine, and he will complete courses in the UNC School of Public Health that pertain to clinical epidemiology, study design, data analysis, and the responsible conduct of patient-oriented research. In addition, he will complete a 5-year research project under the mentorship of Dr. Tim Carey. The primary objective of the study is to develop predictive models for one-year survival and functional status. Secondary objectives will be to describe long-term patient oriented outcomes including health status and quality of life. Patients who require prolonged mechanical ventilation account for 10 to 20% of ICU patients and consume up to 37% of total intensive care unit (ICU) resources. Survivors are often of advanced age and leave the hospital with numerous unresolved medical problems. Preliminary data indicate that one-year survival is as low as 23%. These patients will increase in number as our aging population continues to encounter advances in life sustaining therapies. It is important to determine if these advanced therapies translate into acceptable long-term outcomes. Specific Aims of this study include: 1) Assess one-year survival and functional status for patients who require prolonged mechanical ventilation after acute illness. 2) Determine Health Status and Quality of Life for survivors. 3) Develop and validate a prediction model for one-year survival. 3) Develop and validate a prediction model for independent functional status. This study will enroll a prospective cohort of 200 adult patients requiring mechanical ventilation for at least 21 days at a large tertiary care medical center. Patients will be followed during hospitalization and through the subsequent year. Telephone interviews will be conducted 3 months, 6 months and 12 months after study enrollment to determine patient survival, health status, and perceived quality of life. Using variables measured on day 21 of mechanical ventilation, predictive models will be developed for one-year survival and independent functional status. Models will be validated prospectively in a cohort of 100 similar patients, and physician estimates of one-year survival and functional status will be compared to model estimates. Knowledge of these outcomes would be helpful to clinicians as they counsel patients and their families regarding appropriate levels of aggressive care and help them understand the types of resources and support that will be needed for optimal recovery upon discharge. In addition, these data will inform future cohort studies and randomized controlled trials designed to improve outcomes for this growing patient population.